© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21041
    Prev
    Next
    INDEX
    Explanations

    CDs, City, Gulf Coast, Sideband, airwaves

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     different
    -1.97
     as
    -1.84
     if
    -1.67
     new
    -1.65
     following
    -1.59
     two
    -1.55
     only
    -1.54
     at
    -1.54
     three
    -1.54
     with
    -1.52
    POSITIVE LOGITS
     dát
    1.55
     hörte
    1.54
     déchir
    1.53
     hoffen
    1.50
    痉
    1.48
    delion
    1.43
     réessayer
    1.43
     этих
    1.41
    ”;
    1.40
    是一种
    1.38
    Activations Density 0.315%

    No Known Activations